On Fri, 27 Dec 2013 12:54:57 -0800, Jon Perryman wrote:

> ... I looked at a couple [dependent functions] from IBM very long ago. They 
> basically used it to change FMID ownership of modules. Modules shipped in 
> PTF's under the parent FMID would be ignored and only the new FMID modules 
> would replace the module. ...
>
This can be controlled only by the FMID() operand of the ++VER of the FUNCTION.
Attempting to do so with a PRE() operand would leave an exposure for MODID
failures when a given element is updated in both the parent and the dependent
functions.

> ... Parent is only loosely used for lack of a better term.
>
As long as there's no better term, "parent" is the best available.  SMP/E might
have made things clearer by devising new syntax such as PARENT() or BASE()
(both implied in the documentation) for this construct.

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to